PDA

View Full Version : L-300 Bypass Capacitor Experiment



Regis
06-08-2005, 08:57 AM
After extensively reading the 4343 crossover thread (more than a few times) and the lengthy 'Capacitor' thread started in 2003, I came to a couple of conclusions. The discussion on Solens, Hovlands, Auricaps, Sonicaps, etc, etc was pretty intriguing. It came down to many of the members having to listen to their mod'd speakers very, very carefully for any nuances that might occur. Some of the mods seemed to jump right out, like the Solen caps adding far too much brightness or the perils of tightly wound wire on capacitor construction.

I wonder how much of audio has become 'black magic' (think Monster cable here)? Does that include caps as well? I'm beginning to think so. While my experience is somewhat limited, my observation of the JBL crossovers from the L-110, L-112, L-150 and L-150A indicates that JBL put pretty good quality components in their networks, but none of it was 'black magic' type of components.

Scott Fitlin had apparently used Xicon Caps to good effect and in going to Mouser Electronics, I found .005 uf Polystyrene film capacitors for 30 cents each. The metallized Polypropylene's are 46 cents each. I ordered 30 of each and I'm going to try them out on one of the L-150's I've modified with the L-112 crossover (mod'd to the 3113b configuration per Giskard). The L-112 crossover has the early .068 bypass caps (later on, JBL went with .01 uf on all their systems)

I downloaded the L-250 schematic and saw the .01 uf/.005 uf bypass configurations on every large cap, including the LF, so I'm going to bypass the LF as well. I'll do an A/B listen to the L-150's and then it will be the L-300's next. The L-150 already has a bypassed crossover, so I'm not expecting a huge improvement as I'm replacing the .068 uf caps with the .01 PP and .005 PS Xicon caps (though there was a tremendous improvement using the L-112 bypassed crossover vs the stock L-150 network, which was unbypassed) and I will bypass the LF cap. The L-300's will be next for an A/B mod/listen as well and I'm pretty sure I'll hear a difference as I have with the L-150's I've mod'd. In my opinion, I don't think you have to spend a ton of money on 'black magic' to make these speakers sound better. They sound very, very good to begin with. The mods any of us perform should be just icing on the cake.

Earl K
06-08-2005, 10:36 AM
Scott Fitlin had apparently used Xicon Caps to good effect and in going to Mouser Electronics, I found .005 uf Polystyrene film capacitors for 30 cents each. The metallized Polypropylene's are 46 cents each. I ordered 30 of each and I'm going to try them out on one of the L-150's I've modified with the L-112 crossover (mod'd to the 3113b configuration per Giskard). The L-112 crossover has the early .068 bypass caps (later on, JBL went with .01 uf on all their systems)

- I wouldn't limit your explorations to just the above mentioned values .

- If you really want to get a handle on what different dielectrics' signatures sound like ( as marketed by any one company ) , then I'd recommend bumping up your combined bypass values to around .5 uf . I know this is certainly approaching the "overdose" area of bypassing ( IME ) / but the advantage of this approach is it offers a faster learning curve. Listen to the results & then start reducing the total uf count to what seems to be just right for your music / your ears / your HiFi equipment & the acoustics that you also must listen to .

- IME , it's impossible to predict the "efficacy" of these things from just reading/knowing capacitor brand names or dielectric types. Be prepared to toss the logic . I now own over a couple hundred bypass caps alone and want more because I like playing around with these things. For instance ; a KP type ( Polypropylene Film & Foil ) made by "STK", has such a defined ( pointed ) articulation signature, that using just .044 uf adds much much more articulation to a circuit ( usually too much ) when compared to using .24uf from a different brand ( F-Dyne in this case ). I find the F-Dyne is a lot "rounder" or softer ( really a lot more like my surplus PolyStyrenes ). I really enjoy having both of these KP types to work with. FWIW; my "F-Dynes" predate the included pdf.

- Have fun,

:cheers:

PS - ( FWIW, you are now dabbling in the "blackarts" :p , like it or not ;) ).

Mr. Widget
06-08-2005, 10:49 AM
After extensively reading the 4343 crossover thread (more than a few times) and the lengthy 'Capacitor' thread started in 2003, I came to more than a couple of confusions.;)

Please do plunge ahead and let us know what you discover.:thmbsup:

Widget

Regis
06-08-2005, 11:14 AM
Being a former electronic technician and certified Aircraft Electrician (tho' I've forgetten more than I know...) and being especially anal about how my speakers sound, I'm approaching this with a skeptical mindset. The only reason I'm trying this is because JBL has embraced this technology as Giskard has noted in other technical threads for over 20 years now. I've seen the difference it's made with the L-150's, yet it's going to be hard to believe that I can make the L-300's sound better. They are such astounding speakers and it's taking a true leap of the imagination to even think they can sound better than they do, though there was a noticable diff with the new diaphragms (not a huge diff, but it's there).

Earl, I truly understand what you're saying about the tons of caps and the effects on the sound that you've noticed, but I'm going from the perspective that I only want to remove the damn heavy 15" 136A's once! My second notion is that I fully agree with Giskard about the $100 cap syndrome. He modified his clients L-300's using JBL caps and his client was very happy. When I listen to the modified single L-300 against the unmodified one, I'm going to be critical of how it sounds and if sounds better, than I'll be plenty happy and I'll leave well enough alone, otherwise, I'll always be looking for the 'ultimate combination' of caps. JBL went with .01uf bypass caps on many of their pro-networks. Go to the link below and check out the "4412A-RL" network, the 4612, the Control 12-SR or the 3155. The all used .01uf bypass caps, so that's what I'm going with (see linkee below). I'm very, very curious as to the LF bypass. Some of the early networks didn't bypass the LF and the L-250 did. I believe it's to properly phase all the drivers together. Also wondering how the 077 will sound bypassed too.

http://www.jblproservice.com/navigation/Network%20Schematics.html

The final notion was cost. It's going to be all of $9.00 to modify the L-300 networks with the Xicon caps. I believe JBL would've went the same route, if they hadn't already done so on the L-250's, after all, profit is profit. So I don't believe I'm fully embracing the 'black arts' here, but instead following JBL philosophy of crossover design as it applies to older networks. And yes, Sir Widget, I will follow up with conclusions and photo's!

rek50
06-08-2005, 01:14 PM
Regis, that's what I like, CHEAP caps, thanks for the lead. The satisifaction of improving your gear (anything-cars,audio,house...everything) IMO is priceless. Along the lines of "It's NOT what you can buy that counts, It's what you can BUILD, that counts". Mo Hotta, Mo Betta...

spkrman57
06-08-2005, 03:27 PM
I have found some speakers to react adversely to better quality caps.

Sometimes it is the marriage of the proper crossover caps to the drivers that mate up to them better.

I have found out it doesn't always play better with better cap, there are some exceptions out there.

Of course, I found all this out more my accident, then knowing what I was doing.

Ron

Mr. Widget
06-08-2005, 03:50 PM
Of course, I found all this out more my accident, then knowing what I was doing.

We call it trial and error... I think the big boys use this method to.:)

Widget

scott fitlin
06-08-2005, 05:32 PM
Its Black Magic balancing for sure!

Some electronics and speakers respond favorably to cheapie caps, and adversely to expensive well made, tight tolerance caps!

And I have components that were designed using Premium grade components that sound outrageously clear and revealing of subtle nuance and tonal shading, yet still maintain warmth and presence!

Try what you may, and see what you like!

:D

Earl K
06-08-2005, 05:52 PM
Being a former electronic technician and certified Aircraft Electrician (tho' I've forgetten more than I know...) and being especially anal about how my speakers sound, I'm approaching this with a skeptical mindset.

- That's a good mindset to have with these questions .

- For instance ;
(i) the turnover frequency ( F3 ) for a .01uf "bypass" cap into an 8 ohm load is @ 1,990.Khz ( or just shy of 2 million hz )
(ii) for a .47uf "bypass" cap into the same load its' approx. 42.3Khz ( or 42,328. hz )

Obviously from figures like those , it's not the extra UHF information ( in the form of extra, measurable db gain ) that one hears / IME / it's the restoration of the leading edge of the transients that is heard or sensed .


The final notion was cost. It's going to be all of $9.00 to modify the L-300 networks with the Xicon caps.

- Hopefully this will give you what you want .
- My purpose for including the opposing experiences obtained with 2 different industrial brand caps was to underscore the folly in thinking that all Film & Foil Polypropylenes ( of similar value ) will give the same results . I know this not to be true . The "softer" sounding FFP(s) ,IME, will necessitate the use of higher values to achieve anything close to a "sharp" focus / and still, it's not the same / not better or worse, just different .
- From all the anecdotal evidence ( various testamonials ), the .01uf AudioCap Thetas do appear to offer lot of bang for the buck ( in adding definition or "sharpening the focus") / though I have yet to purchase any or listen to them .


but I'm going from the perspective that I only want to remove the damn heavy 15" 136A's once!
- I hear you, this is all a real PITA / that's why I keep my networks external .

:cheers:

Regis
06-17-2005, 09:31 PM
I completed this mod last weekend. The L-300 crossover is a unique little beast. Not your typical crossover. I was used to circuit boards, with nice traces wandering about. I had an idea what I was getting into, but once into it, it was somewhat daunting. What I'm talking about is that the L-300 is a 'point-to-point' wired device, somewhat like a piece of industrial machinery and uses terminal blocks for the wiring. The schematic is helpful, but it's better to make that direct relationship between the wiring and the terminals.

First off, I didn't remove the entire crossover and associated L-pads out of the cab, I kept it simple '"necessity may be the mother of invention, but laziness is the father'. After removing the 15" woofer, I pulled the red and black input wires from the back of the cab. I then removed the four screws holding the screen down and found that you could simply push the two wire harnesses coming out of the back out of the holes cut in the screen (there is a plastic ring to protect the wires that snaps in, but it pushes right off the side). This gives you extra slack so you can pull the box all the way out of the front of the cabinet.
Find a cardboard box about four inches high and it'll rest perfectly without straining the two harnesses coming off the top. I had a large piece of clean cardboard on the floor to set my soldering iron, magnifying lamps and tools on.

Regis
06-17-2005, 09:33 PM
The box coming on out...

Regis
06-17-2005, 09:37 PM
The next step is to prepare the bypass caps. I directly soldered the .005 uf Xicon Polystyrenes leads to the leads of the .01 uf Polypropylene (this is a parallel circuit...)

Regis
06-17-2005, 09:45 PM
The tricky part was trying to figure out how to bypass the two LF and one large MF caps underneath the terminal strip board. These are embedded in resin and because they're a 'point-to-point' direct electrical set-up (unlike the nicely laid-out circuit board I mentioned earlier) it took some real head scratching to figure out what parallel caps were going to go where.

The big cap in the middle actually had the wires going to opposite corners of the box, so to bypass that, I used an empty terminal strip for this one and jumpered the wires from their point of termination to the empty strip. From the bottom, if you count six blocks, you'll find the first bypass and I outlined this in red. The second cap terminated across the left side from position 4 and 5 and I applied a cap directly across this. The last cap terminated at position 4 as well and from the opposite corner, approximately three down from the very top right group of three termination tabs. The pic below makes it easier to look at. I also added dashed lines to the jumper wires I added. The yellow one on the right, I ran under the board, but the others, I just ran out on top.

Regis
06-17-2005, 09:47 PM
a view underneath (kind of limited, because it only comes up an inch and a half or so because of wire).

Regis
06-17-2005, 09:50 PM
The others were way easy. Those are the three right on top of the board across the bottom. A no-brainer.

Regis
06-17-2005, 10:01 PM
It turned out really well. My belief with bypass caps is that it should be icing on the cake. Here are my thoughts on the result. I noticed better detail, with less blending of the mid and mid-bass on some of the instrumentals. Imaging was improved somewhat. Most noticable on both the L-300 and the L-150 is that they both lost some of the 'boominess' in the bass. The Summit responded especially well, with the bass tightening up really well. I'm happy with the results. Don't expect a huge difference, remember, these are very small values and as I said it's icing on the cake, though the speakers are noticably more pleasing to the ear. I found it hard to believe you could improve on the sound of an L-300, but this works nicely. The caps were of good quality, typical of what I believe JBL would've supplied OEM (the bypass caps I pulled off the L-150's "L-112" crossovers were .068 uf, 20% tolerance caps, nothing super fancy here) so that's the route I took. By the way, even empty JBL L-112 cabs can have some use besides looking ugly with big holes. They make nice stereo equipment tables (saved on an entertainment center).

Again, if you like to play with cap combinations, my hat's off to you (Earl K!). Have fun and I do respect those like Ian McKenzie and Zilch who put together some pretty exotic X-overs. My whole mission here was to show a modification at minimun expense and trouble for L-300, 4333 owners. Thanks for reading.

Earl K
06-18-2005, 05:03 AM
Nice Pictorial Regis !

Thanks for the testamonial . :D

Earl K
06-18-2005, 05:12 AM
For those interested in persuing bypass mods like this , here's the link to Mousers' Xicon "MPP" type caps (http://www.mouser.com/?No=0&handler=data.listcategory&Ne=100&terms=Xicon&Ntt=*Xicon*&crc=true&Ns=SField&N=184) .

And heres the link, to Mousers' Polystyrene caps (http://www.mouser.com/?Ne=100&handler=data.listcategory&Ntt=*Xicon*&terms=Xicon&crc=true&N=2310&Ns=SField) .

:p

johnaec
06-18-2005, 07:30 AM
Nice mods! I'm thinking of doing the same to the crossovers in my 4315's. It's got the same kind of crossover can, and I've already determined, (by comparing the schematics with the crossover), that I can do it like you did, using the point-to-point connections in the top. 'Just have to get around to it...

John

Regis
06-18-2005, 08:04 AM
Thanks Earl for the linkee's to Mouser's PDF files, that was a big help. I shoulda thoughta that! For the .01uf metallized polypropylene, I used Xicon (Mouser PN 1429-6103), they're way over-rated at 630 WVDC, but that was the only MPP cap at that value, .01uf didn't exist for the the lower wattage ones (though, I overlooked the 400 WVDC as I look at the PDF... Doh! I could've saved 11 cents apeice darn it!). Finally, these were radial leads (side-by-side) and it would've been better to find an axial lead, but Xicon doesn't make a metallized Polypropylene in that value/voltage. They worked just fine anyway...

The polystyrenes film caps were Xicon (Mouser PN 23PS250). What's nice about dealing with Mouser vs. Digi-Key (although Digi-key has great people and does have almost everything you need, they didn't have these caps, believe it or not), there is no minimum order (Digi-Key charges an extra $5 or so).

John, good luck with the 4315's! It's not that hard to do. The first one is pretty tough, as you want to be careful, but the second one is a breeze. You'll be happy with the results and it does make a difference (not a huge diff, but a diff nonetheless). Don't forget to pre-tin the empty terminal tabs on both sides of each tab (it makes the soldering of the jumper wires and caps much easier).
:D

rek50
06-18-2005, 09:08 AM
:applaud: Good Show!